1 /*
2  * Copyright (c) 2021 Omar Polo <op@omarpolo.com>
3  *
4  * Permission to use, copy, modify, and distribute this software for any
5  * purpose with or without fee is hereby granted, provided that the above
6  * copyright notice and this permission notice appear in all copies.
7  *
8  * THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
9  * WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
10  * MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
11  * ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
12  * WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
13  * ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
14  * OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
15  */
16 
17 #include "compat.h"
18 
19 #include <stdlib.h>
20 #include <string.h>
21 
22 #include "parser.h"
23 #include "telescope.h"
24 #include "ui.h"
25 
26 /*
27  * Load a text/gemini page given the string page.  Always returns 0.
28  */
29 int
30 load_page_from_str(struct tab *tab, const char *page)
31 {
32 	parser_init(tab, gemtext_initparser);
33 	if (!tab->buffer.page.parse(&tab->buffer.page, page, strlen(page)))
34 		abort();
35 	if (!tab->buffer.page.free(&tab->buffer.page))
36 		abort();
37 	ui_on_tab_refresh(tab);
38 	ui_on_tab_loaded(tab);
39 	return 0;
40 }
41 
42 void
43 parser_init(struct tab *tab, parserfn fn)
44 {
45 	erase_buffer(&tab->buffer);
46 	fn(&tab->buffer.page);
47 	tab->buffer.page.init = fn;
48 }
49 
50 int
51 parser_parse(struct tab *tab, const char *chunk, size_t len)
52 {
53 	return tab->buffer.page.parse(&tab->buffer.page, chunk, len);
54 }
55 
56 int
57 parser_free(struct tab *tab)
58 {
59 	int	 r;
60 	char	*tilde, *slash;
61 
62 	r = tab->buffer.page.free(&tab->buffer.page);
63 
64 	if (*tab->buffer.page.title != '\0')
65 		return r;
66 
67 	/*
68 	 * heuristic: see if there is a "tilde user" and use that as
69 	 * page title, using the full domain name as fallback.
70 	 */
71 	if ((tilde = strstr(tab->hist_cur->h, "/~")) != NULL) {
72 		strlcpy(tab->buffer.page.title, tilde+1,
73 		    sizeof(tab->buffer.page.title));
74 
75 		if ((slash = strchr(tab->buffer.page.title, '/')) != NULL)
76 			*slash = '\0';
77 	} else
78 		strlcpy(tab->buffer.page.title, tab->uri.host,
79 		    sizeof(tab->buffer.page.title));
80 
81 	return r;
82 }
83 
84 int
85 parser_serialize(struct tab *tab, struct evbuffer *evb)
86 {
87 	struct line	*line;
88 	const char	*text;
89 	int		 r;
90 
91 	if (tab->buffer.page.serialize != NULL)
92 		return tab->buffer.page.serialize(&tab->buffer.page, evb);
93 
94 	/* a default implementation good enough for plain text */
95 	TAILQ_FOREACH(line, &tab->buffer.page.head, lines) {
96 		if ((text = line->line) == NULL)
97 			text = "";
98 
99 		r = evbuffer_add_printf(evb, "%s\n", text);
100 		if (r == -1)
101 			return 0;
102 	}
103 
104 	return 1;
105 }
106 
107 int
108 parser_append(struct parser *p, const char *buf, size_t len)
109 {
110 	size_t newlen;
111 	char *t;
112 
113 	newlen = len + p->len;
114 	if ((t = calloc(1, newlen)) == NULL)
115 		return 0;
116 	memcpy(t, p->buf, p->len);
117 	memcpy(t + p->len, buf, len);
118 	free(p->buf);
119 	p->buf = t;
120 	p->len = newlen;
121 	return 1;
122 }
123 
124 int
125 parser_set_buf(struct parser *p, const char *buf, size_t len)
126 {
127 	char *tmp;
128 
129 	if (len == 0) {
130 		p->len = 0;
131 		free(p->buf);
132 		p->buf = NULL;
133 		return 1;
134 	}
135 
136 	/*
137 	 * p->buf and buf can (and probably almost always will)
138 	 * overlap!
139 	 */
140 
141 	if ((tmp = calloc(1, len)) == NULL)
142 		return 0;
143 	memcpy(tmp, buf, len);
144 	free(p->buf);
145 	p->buf = tmp;
146 	p->len = len;
147 	return 1;
148 }
149 
150 int
151 parser_foreach_line(struct parser *p, const char *buf, size_t size,
152     parsechunkfn fn)
153 {
154 	char		*b, *e;
155 	unsigned int	 ch;
156 	size_t		 i, l, len;
157 
158 	if (!parser_append(p, buf, size))
159 		return 0;
160 	b = p->buf;
161 	len = p->len;
162 
163 	if (!(p->flags & PARSER_IN_BODY) && len < 3)
164 		return 1;
165 
166 	if (!(p->flags & PARSER_IN_BODY)) {
167 		p->flags |= PARSER_IN_BODY;
168 
169 		/*
170 		 * drop the BOM: only UTF-8 is supported, and there
171 		 * it's useless; some editors may still add one
172 		 * though.
173 		 */
174 		if (memmem(b, len, "\xEF\xBB\xBF", 3) == b) {
175 			b += 3;
176 			len -= 3;
177 		}
178 	}
179 
180 	/* drop every "funny" ASCII character */
181 	for (i = 0; i < len; ) {
182 		ch = b[i];
183 		if ((ch >= ' ' || ch == '\n' || ch == '\t')
184 		    && ch != 127) { /* del */
185 			++i;
186 			continue;
187 		}
188 		memmove(&b[i], &b[i+1], len - i - 1);
189 		len--;
190 	}
191 
192 	while (len > 0) {
193 		if ((e = memmem((char*)b, len, "\n", 1)) == NULL)
194 			break;
195 		l = e - b;
196 
197 		if (!fn(p, b, l))
198 			return 0;
199 
200 		len -= l;
201 		b += l;
202 
203 		if (len > 0) {
204 			/* skip \n */
205 			len--;
206 			b++;
207 		}
208 	}
209 
210 	return parser_set_buf(p, b, len);
211 }